The objective of the proposed work is to elucidate the relationship between expression of pathobiologically relevant surface components of human prostatic carcinoma cells and tumor cell behavior. The appropriate monoclonal antibodies will be used to detect the expression of specific cell surfact components, as selective agents to enrich for cells with or without a given surface component, and as reagents to simulate host factors which may modulate expression of surface components. A panel of human prostate carcinoma cell clones, derived from established lines, which are positive or negative with respect to surface expression of each determinant detected by the monoclonal antibodies will be established. Cell clones which are positive or negative with respect to expression of each cell surface component under study will be compared in terms of tumorigenicity, growth rate, and metastatic behavior in athymic nude mice. These data will be supplemented by identification of factors which regulate or alter expression of any prostate tumor related cell surface component whose presence or absence results in alteration of in vivo or in vitro characteristics of th parental tumor cell line. These factors will include the influence of specific antibody on antigen expression, the relationship between cell cycle phases and cell surface component detectability, and the role, if any, of cell surface component "shedding" in cell surface expression. Identification of the correlation(s) between the presence or absence of one or more prostate tumor cell related cell surface components, defined within the context of normal surface components and other prostate tumor related components, and biological behavior of the cells in the nude mouse will premit identification of specific molecular entities involved in prostate tumor growth and metastasis.